Project Summary/Abstract Suicide is a grave public health issue. More than 40 thousand people die by suicide in the US annually, making it the 10th leading cause of death1 and responsible for >$30 billion due to lost productivity and medical costs.2,3 For every victim of suicide there are approximately 250 more who engage in non-lethal suicidal behavior, which also contributes significantly to the disease burden. Over the past several decades researchers have identified many risk factors for suicidal behavior; however, the field is no better at predicting or preventing suicide than 50 years ago.4 One way that psychological science can better predict and ultimately prevent suicide is by understanding the processes through which risk factors operate. One psychological mechanism that may confer suicide risk is dysfunctional decision-making. Notably, little research has appropriately examined the decision-making processes of people who make the decision to end their lives. The proposed project addresses this gap by leveraging advances in decision science and computational modeling to more accurately simulate the conditions that engage decision-making processes believed to underlie decisions to attempt suicide. Specifically, because both prevailing theoretical work7?9 and empirical evidence10 suggest that individuals engage in suicidal behavior to escape from aversive stimuli (a negative reinforcement process), the proposed study uses a negative reinforcement based learning paradigm to test people's ability to learn to make decisions that provide relief from an aversive context (i.e., unpleasant noise) in the presence of suicide-related stimuli. Thus this novel behavioral method taps a crucial capacity putatively involved in suicidal decisions: i.e., the capacity to make adaptive decisions to gain relief from aversive experience. Results from a pilot task are extremely promising. When making decisions to gain relief, suicidal individuals demonstrate a specific decision-making deficit relative to controls in the context of suicide-related stimuli but not positively-valenced, matched stimuli (OR=-3.5, p <.05).13 This decision-making deficit may represent a novel marker of suicide risk. The goal of the proposed study is to further examine the association between decision-making and suicide by addressing the limitations of our pilot study and expanding its reach. The proposed study's greatest potential impacts include establishing an evidence base for a novel behavioral marker of suicide risk and gaining a mechanistic understanding via computational modeling techniques of relevant decision-making processes that may confer specific risk for suicidal behavior. If successful, future work based on the proposed study may one day translate findings from this brief and highly scalable procedure into better prediction and therefore prevention of suicide as well as provide novel targets for clinical and/or neurological intervention.